#rag

基于七牛云产品构建的 Model Context Protocol (MCP) Server，支持用户在 AI 大模型客户端的上下文中通过该 MCP Server 来访问七牛云存储资源、利用 Dora 服务进行图片操作等。 如果有什么需求欢迎在下方评论，您也可以在 github 仓库中提 issue。

ContextBridge (CB) is a local-first retrieval layer that sits between your codebase and your AI coding agent. Instead of pasting raw source files into context or letting the AI guess which files matter, CB indexes your codebase's real structure (via Graphify) and returns a compact, ranked result — owner file, related files, key symbols, and dependency chains — grounded in your actual code. A typical response is ~4–8 KB instead of the 100+ KB of raw source an AI would otherwise need to read to answer the same question — roughly a 96% cut in input tokens sent to your cloud AI, with zero hallucinated file paths or method names.

Local-first code retrieval for AI agents — cuts codebase context from tens of thousands of tokens to a few hundred, with zero hallucinated file paths.

Private memory layer — one library your notes, docs, and chat history live in, that any AI tool (Claude, Cursor, ChatGPT) connects to over MCP instead of re-pasting context. Semantic search with citations; every memory is a document you can read, edit, and delete. ~80% fewer context tokens than pasting your library. OAuth sign-in (Claude.ai/Desktop) or API key — get started at mdenh.dev.

LLMs give wrong answers from documents because the extracted text is broken: split words, lost tables, page furniture, and scanned pages that come out empty. PackForAI turns PDF, DOCX, PPTX, XLSX, CSV and JSON into clean, AI-ready Markdown — rejoining words, rebuilding reading order, recovering scanned pages with OCR, and stripping noise. You get compact Markdown plus a before/after token count (typically up to 65% smaller). Use it from the web app, the REST API, an open-source MCP server (Claude & Cursor), or the CLI. Files are stored in the EU, private to you, and never used to train AI.

Upload files programmatically and get back a permanent, shareable URL. REST API and MCP server for AI agents. Requires a FileToUrl Developer plan (£48/year).

AI-calibrated document to Markdown converter. Get your free API key at docpurify.com/mcp. Supports PDF, Office docs, images with structure-aware extraction, AI calibration, table preservation, and smart cleanup.

Parse PDF / Word / PPT / HTML into clean Markdown: doc_parse (page ranges), doc_tables (tables as JSON), doc_images (extract embedded images), doc_chunk (heading-based RAG chunks).

Zero-auth hosted MCP for AI builders — 22 tools: image optimize/convert (WebP/AVIF/JPEG), QR, cited Cleanor Labs storage & image-format data, plus 18 deterministic dev utilities (hash, HMAC, UUID, JWT, JSON, regex, cron, datetime, color, diff…) LLMs get wrong. No API key.

Connect your portfolio to Claude, ChatGPT, or Codex to analyze it and make smarter investments.

An MCP server that gives any LLM or agent clean YouTube transcripts on demand: a single video, a whole channel, or a playlist, plus AI cleanup of auto-generated captions. API-key auth, credit-based, same backend as the public v1 API. Get a free API key with 25 free credits at youtubetranscriptdownload.com/account.

Pay-per-call crypto market intelligence for AI trading agents — 20 tools: price, funding rate, market regime, orderbook, volatility, momentum, Fear & Greed, top movers, gas, stablecoin pegs, ENS, one-call market summary, plus unique Japan data (bitFlyer/Coincheck JPY board, Japan premium arbitrage signal, JPYC yen-stablecoin peg). Remote streamable-HTTP MCP at https://cryptojp.com/mcp (free), with an x402 paid API settling in USDC on Base and JPYC (yen) on Polygon. No API keys, no signup.

memo is a local-first, long-term semantic memory for AI coding agents (Claude Code, Cursor, Cline, Codex). Runs fully offline with in-process embeddings — MLX on Apple Silicon or CPU sentence-transformers on Linux/Docker — and a sqlite-vec + BM25 hybrid retrieval pipeline with reranking. Markdown files on disk are the source of truth; the SQLite index is a rebuildable derivative. Includes a recall hook that auto-injects relevant memories every prompt, an MCP server (memo-mcp), a CLI (memo), a knowledge graph, temporal/time-machine queries, contradiction detection and nightly synthesis. No cloud, no API keys. Install: pipx install mlx-memo

Chat with any product's documentation from Claude, Cursor, Copilot, or any MCP client. Biel.ai turns technical docs into a queryable MCP server with accurate, sourced answers.

Unified prediction-market data API + remote MCP server. Normalizes Kalshi and Polymarket into one schema and surfaces the same real-world event priced on both venues, with the spread between them (cross-venue arbitrage signal). Three tools: list_markets, get_market, matching_markets. Remote streamable-HTTP — no install: claude mcp add --transport http bellwether https://api.bellwether.markets/mcp. Free tier 120 req/hr, no signup.

Real-time Google, Bing, Yandex, and DuckDuckGo SERP data for AI agents, RAG workflows, SEO monitoring, and market intelligence.

Kvasir Legal — verifiable law for AI agents, now with interactive apps Kvasir is a grounding layer for legal AI: 439,000+ statutes and 43,000+ court decisions across six jurisdictions — Germany (federal, Bavaria, NRW), Austria, Switzerland and the EU — every object carrying its provenance (official source, content version, retrieval date). Six tools: resolve_citation (free-text citation → canonical object with pinpoint paragraph + deeplink), verify_citations (batch hallucination check, ≤200 citations per call), search_legal (semantic + full-text), get_legal_object / get_legal_objects, legal_uri_grammar. New — interactive MCP Apps (Claude): citation checks render as a traffic-light report directly in the chat (every citation green/red with source and deeplink, one-click fix suggestions), and resolved statutes render as formatted norm cards with the exact wording and provenance badge. Connect: https://kvasir.legal/mcp — OAuth 2.1 (sign in with your kvasir account, no key handling) or X-API-Key header. Works with Claude, ChatGPT (developer mode connectors), Le Chat, Copilot/VS Code, Cursor and any streamable-HTTP MCP client. Try the REST layer without a key: curl "https://kvasir.legal/api/v1/resolve?cite=§ 242 BGB" Docs: https://kvasir.legal/developers · Invite beta: hello@kvasir.legal

Mnemos is a local-first memory for AI agents, shipped as a single Go binary. It indexes your project's notes, docs, ADRs, source, and OKF knowledge base, then exposes them over MCP so Claude Code (or any MCP client) can search, read, and cite your own knowledge instead of guessing, with no external services (no Ollama, Qdrant, Chroma, Python, Node, or mandatory Docker).

Self-hosted long-term memory for AI agents. An MCP server that gives agents persistent, hierarchical recall (L0–L3) over pgvector — retain preferences, decisions and observations, then recall them by topic across sessions. MIT, clone-and-run, requires Postgres + pgvector.

Agent-first image hosting — upload images (base64 or URL), list, fetch, and delete; returns instant CDN URLs.

BrainTube turns everything you save — YouTube, articles, podcasts, PDFs, Notion, Obsidian, and any URL — into a persistent, semantically searchable knowledge base your AI can query. Connect it to Claude, Cursor, Windsurf, ChatGPT, or any MCP client and search by meaning, not keywords. 52 tools, AI-agnostic, hosted.

FlexOrch MCP converts unstructured documents (PDF, DOCX, invoices, contracts, payroll) into structured, LLM-ready datasets via 6 async tools. Covers the full pipeline: classify → extract fields → mask PII → export JSONL or RAG chunks, with PII detection across 10+ locales (TR, DE, FR, US, UK and more).

Cloud file relay for AI agents: chunked uploads, downloads, folders, encrypted share links, file_read (read text files inline, no download round-trip) and zip_files. 23 tools.

A hosted MCP server that turns PDFs into clean, LLM-ready Markdown, so AI agents can convert documents as a built-in tool: real Markdown tables, formulas as LaTeX, and OCR for scans. The same engine is also a REST API, a web app, a Chrome extension and a ChatGPT Custom GPT. It runs its own engines (MinerU + Docling), not an LLM wrapper. Connect at https://pdf2md.dev/api/v2/mcp with a Bearer API key; docs at https://pdf2md.dev/developers/#mcp

An MCP server that resolves under-specified, half-remembered queries for AI agents into verified answers — or an honest decline. Built on the principle that it never returns a confident-wrong: when it can't resolve with confidence, it shortlists honestly or declines rather than guessing. Free remote endpoint, streamable HTTP transport.

Local-first persistent memory for AI coding agents over MCP. Gives Claude Code, Cursor and Codex shared, durable project memory across sessions. No cloud, no API keys.

Durable, local-first memory for AI coding agents over MCP — zero-dependency, curated & semantically de-duped, you own the data (SQLite + Markdown). Works with Claude Code, Codex & any MCP host.

Agent-native search engine with live web research optimized for AI agents.

Free AI research/OCR + agent-economy discovery tools (120+ tools, no-auth), includes engage_sasame to commission SaSame to build MCP/Claude/RAG/agents, full deliverables pay-per-call via x402

About AIKD (AI Knowledge Daemon) is an MCP tool provider written in Rust that gives AI agents instant access to your codebase. It indexes project files into a searchable knowledge base and exposes 7 tools via the Model Context Protocol (MCP) that any AI agent can call. Use Cases AI-assisted development: Let Claude, Cursor, Cline, or any MCP-capable AI search your codebase semantically Code navigation: Find functions, patterns, and logic across large projects in milliseconds Knowledge management: Index documentation, configs, and code into a unified searchable knowledge base Conversation memory: Save and recall conversation context across sessions CI/CD integration: Use CLI or REST API for automated code search in pipelines

WordPress plugin with a built-in MCP server (remote HTTP, JSON-RPC over POST). Exposes 5 tools: search_knowledge, list_conversations, get_conversation, send_message, get_site_info. Setup: install the plugin from WordPress.org, enable the MCP server in settings, and generate an MCP API key (used as the Bearer token). The server is disabled by default. Full connection guide: https://github.com/rapls/rapls-ai-chatbot/blob/main/MCP.md

TitanStore provides AI agents with programmatic access to compute credits, API keys, cloud storage, and GPU capacity. Search products, manage cart, and complete purchases in a single agentic workflow. No authentication required.

Generates high-quality unit tests for Python code with a coverage before/after report on every run. Derives concrete edge cases (None inputs, numeric boundaries, empty collections, exception paths) and returns ready-to-run pytest files with external I/O automatically mocked. Free tier: 100 executions. Pro: $19/month.

LLMtoMD turns your docs into clean, AI-ready Markdown and serves them to Cursor, Claude Code, and any MCP client so your coding agent retrieves your spec instead of forgetting it.

Documentation and assembly review for @nhtio/adk.

Puter MCP lets your AI tools (Claude Code, Codex, or any other MCP-compatible client) interact with your Puter resources: managing files, publishing websites, deploying workers, and more.

Live crypto market data — price, funding rates, open interest, long/short ratio, order book and candles — across major exchanges, with built-in funding-rate arbitrage comparison.

SCOUTS-AI MCP server: drop-in web_search tool backed by the no-key SCOUTS-AI API (https://scouts-ai.com). For AI agents, RAG pipelines, answer engines, GEO workflows.

A data firewall between AI and your Google Drive. Give Claude, ChatGPT, and other AI assistants controlled access to search, read, create, copy, rename, move, share, and delete files and folders. Sensitive data is redacted before the AI sees it, the AI is held to the permissions you grant, every action is logged, and you can revoke access in one click.

Sifter extracts structured, typed records from your documents (PDFs, scans, contracts, invoices) using a natural-language field spec, then lets an agent query and aggregate them — exact counts, sums, filters, with citations back to the source page. Unlike RAG, it answers collection-wide questions, not just "find the passage.

Chronicle MCP is a high-performance, universal local chat history archive connector designed specifically for Model Context Protocol (MCP) clients like Cursor, Claude Code, and daily development environments. Built natively in Python, it serves as an intelligent local storage and indexing layer that bridges your historical workspace transcripts directly into your active LLM context window. The tool addresses a critical bottleneck in AI-assisted engineering: context window bloat. By locally cleaning, formatting, caching, and dynamically chunking imported historical data, Chronicle MCP reduces context token usage by up to 40% while keeping your assistant fully informed of past decisions, codebase iterations, and project parameters.

MCP server for Yonatan Gross (yonyon.ai) — ask about his work, browse his AI/RAG/agent projects, and book a free intro call. Public, no auth. Tools: ask_yonatan, browse_projects, book_intro_call.

MemClaw — persistent memory for AI agent fleets (OSS)

MCP server for Yonatan Gross (yonyon.ai) — ask about his work, browse his AI/RAG/agent projects, and book a free intro call. Public, no auth. Tools: ask_yonatan, browse_projects, book_intro_call.

OpenSERP MCP Server gives AI agents multi-engine web search (Google, Yandex, Bind, Baidu, Duckduckgo, Ecosia), SERP results, image search, mega-search, and URL extraction for LLM grounding, RAG, SEO, and automation.

Auditable, self-improving knowledge & memory for AI agents, served over MCP. Citation-enforced answers (no source, no claim), a replayable why-trace per answer, and a judge-gated loop that promotes validated knowledge. Zero-key local embeddings (bge-m3), hybrid vector + BM25 retrieval, scoped fail-closed bearer auth, and multi-tenant row-level security — self-hosted on Postgres alone. Apache-2.0. Reference implementation of the Agentic Product Standard.

Turn any published podcast into clean Markdown with real speaker names — not "Speaker 1." Three tools: search_podcasts (by text query or a pasted Spotify/YouTube URL), get_transcript (named, timestamped Markdown ready for LLMs and RAG), and get_balance. It's a transcript-retrieval API, not speech-to-text: it works on already-published shows, so you skip uploading audio and diarization — typically 5–10× cheaper. Try it free with the demo key pt_demo (no signup).

Neural-OS — The mobile-first AI agent OS. Featuring sovereign agent tools for DePIN compute arbitrage, data bounty hunting, enterprise lead bartering, and live crypto pricing execution.

130+ QA & dev tools for AI agents. Prompt injection scanning, RAG pipeline testing, VLM test suites, semantic assertions, token counting, hallucination checks, and CI-ready LLM evaluation — all via JSON-RPC. Free, no API key, no signup.

FinanceGenius.AI is India's first AI-native financial product MCP server. 15 read-only tools that any MCP-compatible AI client (Claude Desktop, Cursor, Continue, etc.) can use to: • Search Indian credit cards, personal/home/car loans, fixed deposits, mutual funds, and insurance from major Indian institutions • Compare 2-4 products side-by-side with diff tables • Run calculators: EMI (with amortization), SIP, CAGR, FD maturity, Indian income tax (FY 2025-26, old vs new regime) • Get live NSE/BSE stock quotes by symbol or popular stocks dashboard • Look up AMFI mutual fund NAVs by scheme code • Get plain-language explainers of Indian financial concepts via hybrid-RAG over product brochures + regulatory guides + live product DB • Find the best FD by net-of-tax yield for a given tenure + tax slab • Find the best credit card for a given spending pattern Free, no auth, rate-limited at the edge. Bias-disclosed: FinMall.AI earns the same trail commission across all products listed; rankings are user-economics first, no pay-to-rank. Powered by FinMall.AI — India's AI-native financial marketplace.

Forge is Voxell's hosted text-embedding API. This MCP server exposes two tools — embed (turn text into vectors) and list_models — so any MCP-compatible agent can generate embeddings for semantic search, RAG, clustering, and dedup. Three tiers let you dial quality against cost: turbo (1024d, fast and cheap), pro (2560d), and ultra (4096d — Qwen3-Embedding-8B, ~75+ average task score on MTEB and currently #4 on MTEB English; the models above it are research-only, so it's the top usable model). Matryoshka dimensions let you truncate to fewer dims — re-normalized — for cheaper vectors. Requires a FORGE_API_KEY (free tier at dash.voxell.ai). Embeddings only — no data is stored.

MCP server for Calypso’s multimodal RAG layer. Supports agent-store uploads, knowledge-store ingestion, and retrieval-backed responses workflows with file-aware rag_policy semantics.

Launch a hosted Gemini File Search true multimodal RAG layer that answers from your PDFs, docs, screenshots, charts, diagrams, help center, FAQs, and images with citations people can verify.

Self-hosted AI workflow automation platform with visual canvas, agents, RAG, HITL, MCP, and observability in one runtime. Expose your Heym workflows as an MCP server at /api/mcp/sse — callable from Claude Desktop, Cursor, or any MCP client.

A local encrypted execution layer for AI agents. Provides sealed scripts, protected subprocess execution, and secure local storage.

Real-time crypto spread, funding rates, and cash & carry yields across Binance and OKX. Built for AI assistants that need live market data.

Doxa MCP handles any question with grace and truth. Any human situation, from Bible lookup and Christian encouragement to apologetics, grief, religious trauma, and doubt, gets a thoughtful, theologically-grounded response in any MCP-aware agent. The voice is hardened against AI-companion patterns, sycophancy, prosperity-gospel framing, and prophecy-fulfillment claims. It points to Jesus, never to itself. 31,000+ Berean Standard Bible verses (public domain), 1,800+ verified testimonies, and ready answers for the hard questions: suffering, science and faith, the exclusivity of Christ, doubt. 5/5 on the independent faith.tools rubric for Christian AI builders. Three tools: doxa_encourage, doxa_scripture, doxa_way_movement. Free hosted, BYOL Anthropic for unlimited. Scripture, summoned.

MEMANTO is a memory agent. It remembers, recalls, and answers - so your agents can achieve long-term goals and avoid confusion.

Drillr is the official financial MCP for AI agents. One hosted endpoint exposes **9 tools** spanning standardized financial data (90+ SQL tables: statements, ratios, earnings, insider, ownership), paragraph-level SEC filing semantic search across 10-K/10-Q/20-F/6-K/S-1/DEF 14A, live cross-asset signals across ~6,900 tickers, and 24 alt-data categories spanning the AI value chain (energy, chips, compute pricing, model adoption, app usage). Coverage: US + Japan equities plus ADRs of global companies; fundamentals back to the 1980s. Built and maintained by Little Grebe Inc.

Ad-free web scraping and search exposed as 3 MCP tools fetch_page, fetch_pages_batch, search_web. Works with Claude Desktop, Cursor, Cline.

Your company's brain for AI agents. Cited, permission-aware knowledge across every connected system — Drive, SharePoint, Confluence, Slack, Notion and more. Reasons over time, authority, and contradiction. Refuses on weak context.

Four specialized AI agent skills: Integral Hermeneutics (8-lens text interpretation), Rune Wisdom (Elder Futhark divination with 14-layer stack), What-If Scenario Builder (15 foresight frameworks), and Strategical Alignment / White Rock (6-phase short-term vs long-term misalignment detection protocol).

Sovereign E2EE cloud storage for AI agents. MCP-native, zero-knowledge, RGPD-compliant, hosted in France.

The open retrieval layer for AI coding agents. Indexes code, docs, legal, research, data — 15 parsers, FTS5 + semantic search, knowledge graph. Serves surgical context via MCP. Open source, local, free. Index everything in your project — code, docs, PDFs, legal texts, research, data — and your agent finds the right context instantly. No hallucinations. No cloud. No API costs.

An MCP (Model Context Protocol) server providing hybrid semantic search over code across a set of GitHub repositories that you list in config.yaml. It parses symbols with Tree-sitter and indexes both code and git commit history, so AI clients can query them by natural language or by symbol name. Hybrid retrieval combines dense embeddings with BM25, so both natural-language queries ("where do we publish order events?") and symbol-name lookups (PlaceOrderRequest) work well.

Open-source MCP memory server for AI agents — persistent, searchable, tiered memory across sessions. Works over stdio (Cursor, Claude Desktop) or HTTP+SSE. MIT licensed (separate commercial license available).

Real-time crypto data for AI assistants. Live price spreads between Binance and OKX, perpetual funding rates, Cash & Carry arbitrage yields, and exchange status. No API key required.

Ground Claude, ChatGPT, Cursor, and Windsurf in 300M scientific works — every citation is a real, verifiable DOI.

Structured knowledge API for AI agents — fractal knowledge graphs, signal datasets, authority evidence reviews, open problems, and capability specs. Pay-per-query or subscription via Lightning, Stripe, or USDC.

The Interplanetary File System (IPFS) MCP toolkit gives AI Agents and automated workflows the ability to utilize IPFS as a backend storage medium. Upload, retrieve, pin, and manage encrypted files on IPFS via a simple API accessible as MCP tools.

Semantic search over 1M+ vectors — annual reports, quarterly filings, press releases, real-time electricity prices and macro context data for Norway, Sweden, Denmark, Iceland and Finland. Free, no API key required.

Open registry of skills, tools, datasets, prompts, and playbooks for local LLMs. Search 155+ resources, get AI-powered recommendations, and install capabilities on demand for Ollama, LM Studio, and llama.cpp. Free, open, community-built.

scratchpad-mcp is an MCP server that gives AI agents persistent, token-efficient storage. It solves a specific waste problem: agents constantly re-read files they've already seen, re-summarize documents they've already processed, and re-load context they've already understood. Every one of those round-trips burns tokens for no new information. This server fixes that with eight tools designed around how agents actually work: Versioned writes. write_file automatically versions every write and keeps the 10 most recent versions per file. Storage is append-only on success and atomic on failure partial writes can't corrupt state. Structured diffs. read_file accepts a since_version parameter and returns a JSON line-diff against that prior version instead of the full content. Agents that have already seen v1 can ask "what changed in v3?" and get a small structured payload they can reason about, not the entire file again. Append-only logs. append_log and read_log give agents an event-stream they can replay. Cursor-based pagination (since_entry + last_entry_id + has_more) means an agent can checkpoint where it left off and resume cheaply. On-demand summaries. summarize_file calls Claude Haiku to summarize files over ~2000 estimated tokens. Summaries are cached per file version, so repeat calls on an unchanged file cost nothing. The threshold is enforced server-side you can't accidentally pay to summarize something small. Per-agent isolation. Every operation is scoped by an agent_id parameter, so one server instance can serve many agents without leaking state between them. Storage limits. 1 MB per file write, 64 KB per log entry, 1000 files / 100k log entries / 100 MB total per agent sane multi-tenant guardrails out of the box. Backed by a single SQLite file (Postgres migration is on the roadmap). All SQL is parameterized, paths are validated against a strict allowlist, and the security model is documented honestly it's safe for one-user-per-process deployments today, and the V2 plan derives agent_id from the caller's API key for true multi-tenancy. Build agents that remember what they've already seen.

Audit-grade memory backbone for agent teams. Bi-temporal facts (event time + transaction time, with recall(as_of=...) replay), 6-step deterministic retrieval (no LLM in the critical path), conversation ingest with speaker-locked dual-pass extraction, per-tenant Postgres row-level security, and Ed25519-signed provenance. Postgres + pgvector + Neo4j defaults; ArangoDB / DynamoDB / Cosmos / AlloyDB also supported. Listed on the official MCP Registry as io.github.bolnet/attestor.

US/HK markets — 133 tools: quotes, options, orders, fundamentals, IPO, alerts, DCA & portfolio

Your Data. Your AI. Your Rules. Ontheia is a self-hosted, open-source AI agent platform. Run AI agents, automate workflows, and connect AI models to any tool — entirely on your own infrastructure, without sending data to external cloud services.

Give your AI agent grounded legal knowledge: search, cite, and read German and EU laws, EU regulations/directives, and federal court decisions — returning verified full text with stable citations.

Give your AI agent grounded legal knowledge: search, cite, and read German and EU laws, EU regulations/directives, and federal court decisions — returning verified full text with stable citations.

Memtrace — Structural Memory for AI Coding Agents The Problem Every AI coding agent — Claude Code, Cursor, Codex, Copilot — starts each turn completely blank. It re-reads raw source files and re-derives the full call graph, type hierarchy, and import tree from scratch on every single invocation. That structural rework burns 60–90% of the context window before any real reasoning begins. Less than 5% of tokens in a typical agentic coding session contribute genuine new intelligence. The rest is expensive, redundant noise — and it compounds: accuracy drops 40% as sessions grow, stale context crowds out signal, and summaries strip out the structural relationships agents need most. The Solution Memtrace is a bi-temporal structural memory layer that turns your codebase into a live, queryable knowledge graph — compiled from the AST, not guessed from embeddings. Every function, class, interface, and API endpoint becomes a typed node with deterministic relationships. Every file save becomes a queryable episode with timestamps, so agents can reason about structure, detect regressions, and time-travel through their own work without re-reading anything. One Rust binary. Zero configuration. Five-minute install. What agents can do with it Find callers, callees, and dependencies instantly — no file scanning, no token waste Compute blast radius before making a change — know exactly what breaks before anything is touched Detect structural drift between sessions — catch regressions the moment they happen, not at PR review Time-travel through code evolution — query any prior state of any symbol, not just git commits Search across the full codebase with hybrid retrieval — BM25 full-text + HNSW vector + graph traversal fused in one query Map API topology across services — cross-repo HTTP call graphs, dependency chains, dead endpoint detection Benefits −90% token cost on structural queries (Mem0) +26% accuracy on multi-step agentic tasks (Mem0) −91% p95 latency on structural lookups vs. RAG baselines +32.8% SWE-bench bug-fix success rate when agents have graph context (RepoGraph) 200–800ms per-save re-indexing — every file save is a queryable episode in under a second 40+ MCP tools covering indexing, search, relationships, impact analysis, temporal evolution, API topology, graph algorithms, and direct Cypher queries 12 languages + 3 IaC formats supported via Tree-sitter grammars Local-first, closed-source Rust — code never leaves the machine, no account required, no telemetry

Search, memory, and blob storage for AI agents. One key, no setup.

AI-first file sharing and collaboration. 251 MCP tools give agents a full workspace: file storage, branded shares, comments, workflows, and built-in RAG. Free tier includes 50GB storage and 5,000 monthly credits, no credit card. Built by the team behind MediaFire. Works with Claude Desktop, Claude Code, Cursor, VS Code, Cline, and any MCP-compatible client. Connect via Streamable HTTP at https://mcp.fast.io/mcp or SSE at https://mcp.fast.io/sse. OAuth 2.0 PKCE authentication. Open source (Apache-2.0).

Japanese LLM security APIs — prompt injection detection (jpi-guard) + PII masking (PII Guard). Free to start, no credit card required.

We design and operate proprietary signal systems that turn noisy market, macro, and media data into structured, calibrated intelligence — exposed through web dashboards, a Signals API, and a public MCP server any AI assistant can reason over directly. The eight pillars Every product above reads from the same nightly signal pipeline (00:05–04:05 UTC) organised into eight analytic pillars: Composite risk monitor — 0–100 regime score with AR / turbulence / GDELT decomposition, extended by a 4-way liquidity gate. Financial turbulence — cross-asset covariance regime with Markov transition probabilities. Systemic fragility — three-tier Absorption Ratio alerts plus per-asset PC1 / PC2 eigenvector history. Media sentiment — GDELT-derived tone, volume, and outliers across 27+ assets plus a global geopolitical tension gauge. Global liquidity — GLI composite with policy / private-sector / cross-border sub-indices and regional breakdown. Statistical arbitrage — 550+ security scanner for oversold and overbought mean-reversion candidates. AI-generated intelligence — cross-pillar briefings and narrative commentary, clearly flagged as synthesized vs deterministic. Supporting analytics — multi-year asset-class returns and a RAG stablecoin scorecard.

Web scraping, crawling, and data extraction for AI agents. 5 built-in tools: scrape (clean markdown from any URL), crawl (follow links across entire sites), map (discover all URLs on a domain), extract (structured JSON via schema), and search (web search with optional page scraping). 833ms average latency, single binary, 6 MB RAM. Self-host free or use managed cloud at fastcrw.com with 500 free credits. Firecrawl-compatible API. Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP-compatible client.

Official MinerU MCP server for document parsing. Convert PDF, DOCX, PPTX, images, and HTML to high-quality Markdown. Supports 109-language OCR, table/formula extraction, and both token-free Flash mode (20 pages / 10MB) and Precision mode (600 pages / 200MB). Works with Claude Desktop, Cursor, and Windsurf.

x402 micropayment API for AI agents. In-memory vector store with cosine similarity search. For RAG pipelines. Pay per call with USDC on Base.

x402 micropayment API for AI agents. Detects funding rate arbitrage opportunities. Pay per call with USDC on Base.

Stop AI hallucinations with certified, source-backed facts. Swiss Truth MCP gives your AI agent access to a curated knowledge base covering Swiss law, health, finance, climate, AI/ML, and more. 8 tools, no API key needed, public and free.

Search real estate listings, development projects, companies, and market data in Paraguay and South America. 8 tools: property search with 20+ filters (city, price, bedrooms, type), project search with unit pricing, company directory, location hierarchy, market statistics, and Airbnb/STR data. Returns Markdown with links to tulugar.com.

MCP server that gives AI coding agents instant access to Thomson MO5 technical documentation via semantic search. Powered by a RAG backend, but zero configuration required: just plug and play with the public instance at retrocomputing-ai.cloud. Part of a complete MO5 development toolchain: - SDK MO5 (https://github.com/thlg057/sdk_mo5): C library optimized for CMOC with MO5-specific abstractions - MO5 Project Template (https://github.com/thlg057/mo5_template): Ready-to-use project scaffold with automated Makefile - This MCP Server (https://github.com/thlg057/mo5-mcp-server): The AI assistant that ties it all together: answers technical questions, searches docs, and generates assets. Ideal for retrocomputing enthusiasts developing in C for the Thomson MO5 with modern agentic tooling.

24 MCP tools for ocean shipping intelligence — D&D tariffs, freight rates, surcharges, local charges, port congestion, vessel schedules, and risk scoring across 11 carriers and 157 countries. Free tier available.

Persistent memory for AI agents, important things stick, forgotten things fade, outdated facts get pruned automatically using the Ebbinghaus forgetting curve.

BitAtlas is a next-generation cloud storage platform designed with privacy and AI agents in mind. It provides a secure, zero-knowledge environment where your most sensitive documents are encrypted client-side and accessible only to you and the agents you authorize.

Persistent project memory for AI coding agents. Hybrid vector + BM25 search with AST-aware chunking (24 languages), dependency graph boosting, conversation memory, checkpoints, annotations, and wiki generation. Local-first, zero config. `bunx mimirs init`

Ocean container shipping intelligence for AI agents: D&D tariffs, freight rates, vessel schedules, port congestion, inland haulage, and total cost calculation across 6 major carriers (Maersk, Hapag-Lloyd, MSC, COSCO, ONE, CMA CGM). 24 MCP tools — 4 free + 20 paid via x402 USDC micropayments.

Cloud object storage with native MCP interface. AI agents like Claude autonomously save and retrieve files via 14 MCP tools. Memory persists across sessions. Web dashboard with file manager, folder sharing, and full-text search. OAuth 2.0 + API Token auth. Free tier: 10GB storage.

Save tokens while coding — your AI agent gets structured code context, not file dumps. CodeWeave is an MCP server that gives AI agents structured code intelligence instead of dumping entire files into context. Your agent queries local indexes — AST, call graph, type graph, and hybrid semantic search — and gets back only what it needs. The result: less token waste, more relevant context, better coding decisions. It supports 7 languages (Python, TypeScript, JavaScript, Go, Rust, Java, C#) with a 6-stage hybrid search pipeline combining vector embeddings, full-text search, and structural density scoring. Everything runs locally with zero external dependencies beyond Ollama for embeddings. Monorepo and git worktree support included. A file watcher keeps indexes current as code changes — no manual reindexing needed. ATTENTION: Run "npx @codeweave/mcp" in your project directory first. The setup wizard handles everything for you. At first, it may take some time. INFO: Exhaustively tested on large Java, C#, JavaScript, TypeScript, React codebases. Contributions and feedback are welcome.

Real-time X/Twitter sentiment intelligence API for AI agents via Grok API + x_search. Automate Social Arbitrage: spot cultural trend shifts, narrative velocity, and contrarian signals BEFORE Wall Street. Returns structured JSON with sentiment scores, narrative velocity, top narratives, KOL positions, and contrarian signals. Pay-per-query 0.005 USDC on Base via x402 micropayments. The affordable alternative to TickerTrends, QuiverQuant, and RavenPack for autonomous trading agents. MCP-native discovery. Free preview endpoint for evaluation.

Munin is a high-performance, pragmatic memory layer for AI agents (Cursor, Claude Code, OpenClaw, Gemini CLI,...). Unlike other solutions, Munin focuses on developer productivity with: - **Multi-Project Support:** Isolate memories into separate "brains" (Context Cores). - **GraphRAG:** Automatically builds a knowledge graph from your context. - **Sub-200ms Search:** Blazing fast Hybrid & Semantic search. - **Privacy:** Zero-knowledge client-side encryption (E2EE) support. - **Pricing:** FREE FOREVER for the base tier (Pro and Elite Plan start from $1.6/m).

Personal YouTube AI knowledge base powered by RAG. Query your subscribed channels.

Trading intelligence tools for Polymarket prediction markets. Slippage estimation, liquidity scanning, arbitrage detection, price feeds, wallet intelligence, and portfolio risk. Built for AI agents.

Make your knowledge agent-ready. Connect docs from Confluence, Notion, GitHub, Dropbox, or Google Drive — any AI agent searches them via one MCP endpoint. 3 retrieval modes: vector search for precision, broad search for exploration, full document access as fallback. The agent decides how deep to dig.

EU AI Act compliance scanning for Python AI agents. 10 tools that scan, analyze, and fix compliance gaps across LangChain, CrewAI, AutoGen, OpenAI, and RAG pipelines. Checks 6 EU AI Act articles with HMAC-SHA256 tamper-evident audit trails. pip install air-blackbox-mcp. August 2026 deadline ready.

Upload, share and transfer files instantly. No signup or API key required. End-to-end encryption, multi-file bundles, free up to 1GB.

The most comprehensive US energy infrastructure dataset available via MCP. 31 tools for Claude, GPT, Cursor, and any MCP-compatible agent. Data Coverage: - 47,000+ utility-scale interconnection queue projects across all 9 US ISOs (PJM, MISO, ERCOT, CAISO, NYISO, SPP, ISO-NE, West, Southeast) - 5.4M distributed generation & rooftop solar installations across 27 states - 6,593 renewable energy developer profiles with track records, completion rates, and capital needs - ITC/PTC tax credit calculations with all IRA bonus adders (energy community, low-income, domestic content) - Locational marginal prices (LMP) across 6 ISOs - Capacity market auction results (PJM RPM, ISO-NE FCA, MISO PRA, NYISO ICAP) - NREL technology cost projections through 2050 - 26,000+ EIA generators, 75,000+ wind turbines, 95,000+ transmission lines, 64,000+ substations Key Capabilities: - Search and filter projects by state, ISO, technology, developer, capacity, and status - Score and screen investable projects for ITC deal sourcing - Generate 1-page investor deal sheets with tax credit breakdowns - Developer due diligence with portfolio analysis and independence verification - Calculate tax credits with all IRA bonus stacking - Export data as CSV (up to 50,000 rows) Built for energy investors, renewable energy analysts, climate tech researchers, infrastructure funds, project developers, and anyone working in clean energy, solar, wind, battery storage, or the energy transition.

Access the most comprehensive US energy infrastructure dataset available via MCP: - 47,000+ utility-scale interconnection queue projects across all 9 ISOs (PJM, MISO, ERCOT, CAISO, NYISO, SPP, ISO-NE, West, Southeast). - 5.4M distributed generation and rooftop solar installations across 27 states. - 6,593 renewable energy developer profiles with track records, completion rates, and capital needs assessment. 31 tools covering: project search and filtering, - ITC/PTC tax credit calculations with all IRA bonus adders (energy community, low-income, domestic content) - investable project scoring for deal sourcing, - 1-page deal sheet generation - developer due diligence - locational marginal prices (LMP) across 6 ISOs - capacity market auction results - NREL technology cost projections - EIA generator data, wind turbine database - transmission lines - substations - fuel prices - CSV data export. Built for energy investors, renewable energy analysts, climate tech researchers, infrastructure funds, project developers, and anyone working in clean energy, solar, wind, battery storage, or the energy transition.

65+ AI tools as an MCP server. Research, write, code, scrape, translate, analyze, agent memory, workflows. Pay per call from $0.006.

Real-time sports card pricing, market analysis, arbitrage detection, grading ROI, investment advice, and player stats (NBA/NFL/MLB). 9 tools for AI agents helping collectors and investors.

Semantic code search for AI agents — hybrid vector + BM25 with cross-encoder reranking, AST-aware chunking for 14 languages, conversation memory, code annotations, and search analytics. Zero config, zero API keys. Just bunx.

Semantic search over IMAGIN.studio API documentation — CDN, API, integration guides, and more.

Real-time sports card pricing, market analysis, arbitrage detection, grading ROI, investment advice, and player stats (NBA/NFL/MLB). 9 tools for AI agents helping collectors and investors.

Semantic search over IMAGIN.studio API documentation — CDN, API, integration guides, and more.

PinRAG is a retrieval-augmented generation (RAG) MCP server built with LangChain and Chroma. Index PDFs, plain text, Discord exports, YouTube transcripts, and GitHub repositories, then ask questions in Cursor, VS Code (Copilot), or any MCP-capable client. Answers include citations (pages, timestamps, paths). Install via PyPI (`pipx`, `uv tool`, or `uvx --from pinrag pinrag-mcp`). Configure API keys in your editor’s MCP `env` block.

A powerful Model Context Protocol (MCP) server using gemini embedding 3 that transforms any local directory into an ultrafast, visually-aware spatial search engine for AI agents.

MCP server for perpetual futures trading across Pacifica (Solana), Hyperliquid (HyperEVM), and Lighter (Ethereum). 18 tools for market data, portfolio management, trade execution with dry-run safety, funding rate arbitrage scanning, and analytics.

Transform 17 source types (docs, GitHub repos, PDFs, videos, Jupyter, Confluence, Notion, Slack/Discord) into AI-ready skills and RAG knowledge. 35 MCP tools for scraping, packaging, and exporting to vector databases. Supports 16+ LLM platforms.

One-line MCP config gives your AI agent persistent memory with automatic workflow rules. Unlike passive memory stores, Awareness teaches your agent WHEN to remember and recall — works across Claude Code, Cursor, Windsurf, Cline, Copilot and 7 more IDEs.

Connect your AI assistant to real-time cryptocurrency market data. Get live prices, OHLCV candlestick history, and trading pair information from Binance, Coinbase, Kraken, OKX, and Bybit. Compare prices across exchanges to spot arbitrage opportunities, analyze price movements, and track any trading pair, all through natural language.

Web fetching, search, and structured data extraction API built for AI agents. Fetch any page as clean markdown, raw HTML, or JSON-LD. Search the web and news across 20+ countries. One credit per request, no tiers, no subscriptions. 500 free requests on signup.